Estimation device and estimation method

Abstract

To accurately estimate a traffic volume on a road in an actual environment.SOLUTION: An estimation device 100 comprises: a probability density estimation unit 152 that estimates a two-dimensional continuous probability density function from a sample distribution for showing a relation between a traffic density and a travel speed for a plurality of vehicles that have previously traveled in a target area; a calculation unit 153 that changes a traffic density which is a continuous variable with respect to the two-dimensional continuous probability density function, and calculates a one-dimensional probability density function of a travel speed corresponding to each traffic density; and a traffic density estimation unit 154 that receives a distribution of travel speeds from the plurality of vehicles traveling in the target area, and estimates a traffic density in the target area based on the one-dimensional probability density function and the distribution of travel speeds.SELECTED DRAWING: Figure 8

Description

[Technical Field]

[0001] The present invention relates to an estimation device and an estimation method. [Background technology]

[0002] While conventional technologies for estimating traffic volume in a target area involve analyzing camera images and videos of moving objects on roads, there is also research underway into traffic volume estimation technology that utilizes speed information collected from vehicles connected to the cloud, as a method that does not require the installation of cameras or other equipment, is relatively low cost, and requires light computational processing.

[0003] For example, as a technology for estimating traffic volume from speed information, there are conventional technologies that use the Underwood model, which is a model between traffic density (K) and travel speed (V), and the Greenberg model.

[0004] In addition, there is a conventional technology that compares a discrete distribution histogram of the speed of vehicles connected to the cloud with a discrete distribution histogram of the driving speed (V) at all traffic densities (K) generated from comprehensive driving data on the relevant road, and determines the traffic density (K) corresponding to the most similar discrete distribution histogram as the final estimated traffic density. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] Tomio Miwa, Toshiyuki Yamamoto, Tomonori Takeshita, Takayuki Morikawa, "Study on the possibility of estimating dynamic OD traffic volume using speed information from probe cars", Journal of the Japan Society of Civil Engineers, Vol. 64 No. 2, 252-265, May 2008 [Non-patent document 2] Osamu Daido, Satoshi Hiromori, Takaaki Umezu, Hirozumi Yamaguchi, Teruo Higashino, "Effectiveness Evaluation of Link Traffic Volume Estimation Method Using Probe Car Data," Information Processing Society of Japan Technical Report, Vol.2012-MBL-64 No.20 Vol.2012-ITS-51 No.20 2012 / 11 / 16 Summary of the Invention [Problem to be solved by the invention]

[0006] However, the above-mentioned conventional technology is unable to estimate traffic volume taking into account factors that have a complex effect on traffic volume, such as road conditions in the real environment, and also has the problem that the estimation accuracy decreases when the amount of speed information collected is small.

[0007] For example, the Underwood model and the Greenberg model only use traffic density (K) and travel speed (V), which represent congested and non-congested conditions on each road, as their main parameters. As a result, they are unable to estimate traffic volume taking into account factors that have a complex effect on the KV relationship, such as the attributes of vehicles traveling on each road, the number of lanes, and weather.

[0008] In addition, when estimating by comparing histograms of discrete variances, the shape of the histogram may change depending on the setting of the histogram bin width, and the estimation result may fluctuate. Furthermore, when there is little observed speed information, it is not possible to evaluate the likelihood of the estimation result.

[0009] The present invention has been made in view of the above, and has an object to provide an estimation device and an estimation method that can accurately estimate the traffic volume of roads in a real environment. [Means for solving the problem]

[0010] In order to solve the above-mentioned problems and achieve the object, the estimation device is characterized by having a probability density estimation unit that estimates a two-dimensional continuous probability density function from a distribution of samples that indicate the relationship between traffic density and traveling speed for multiple vehicles that have previously traveled through a target area; a calculation unit that varies traffic density, which is a continuous variable, for the two-dimensional continuous probability density function and calculates a one-dimensional probability density function of the traveling speed corresponding to each traffic density; and a traffic density estimation unit that receives the distribution of traveling speeds from multiple vehicles traveling through the target area and estimates the traffic density of the target area based on the one-dimensional probability density function and the distribution of traveling speeds. [Effects of the Invention]

[0011] According to the present invention, it is possible to accurately estimate the traffic volume of roads in a real environment. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a traffic volume estimation system according to the first embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of the data structure of the comprehensive data table. [Figure 3] FIG. 3 is a diagram showing an example of a KV scatter diagram. [Figure 4] FIG. 4 is a diagram showing an example of the results of two-dimensional kernel density estimation. [Figure 5] FIG. 5 is a diagram for explaining the process of calculating a one-dimensional probability density function. [Figure 6] FIG. 6 is a diagram illustrating an example of the data structure of the probability density function table. [Figure 7] FIG. 7 is a diagram for explaining the processing in the estimation phase. [Figure 8] FIG. 8 is a functional block diagram showing the configuration of the estimation device. [Figure 9] FIG. 9 is a flowchart showing the processing procedure of the preparation phase. [Figure 10] FIG. 10 is a flowchart showing the processing procedure of the estimation phase. [Figure 11] FIG. 11 is a diagram illustrating an example of a computer that implements the estimation device by executing a program. DETAILED DESCRIPTION OF THE INVENTION

[0013] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of an estimation device and an estimation method according to the present invention will be described in detail with reference to the accompanying drawings. However, the present invention is not limited to the embodiment described below.

[0014] (Traffic volume estimation system) A traffic volume estimation system according to an embodiment will be described below. Fig. 1 is a diagram showing an example of the configuration of a traffic volume estimation system according to an embodiment.

[0015] 1, the traffic volume estimation system 1 includes vehicles 10a, 10b, and 10c, traffic counters 20a, 20b, and 20c, a weather information table 30, and an estimation device 100. The vehicles 10a to 10c, the traffic counters 20a to 20c, the weather information table 30, and the estimation device 100 are connected to one another via a network 5.

[0016] Vehicles 10a to 10c are vehicles equipped with communication devices connectable to network 5, and each transmits "vehicle travel speed information" to estimation device 100. For example, the vehicle travel information transmitted from vehicle 10a includes information such as the average travel speed of vehicle 10a and the position of vehicle 10a. Although only vehicles 10a to 10c are shown in FIG. 1, other vehicles may also be included. In the following description, vehicles 10a to 10c will be collectively referred to as "vehicle 10."

[0017] The traffic counters 20a to 20c are arranged in multiple observation areas and measure the traffic volume (Q) and vehicle traveling speed (V) of vehicles passing through the observation areas. The traffic counters 20a to 20c transmit traffic counter information, which is the measurement result, to the estimation device 100. For example, the traffic counter information includes area identification information that identifies the observation area, the time period in which the measurement was performed, the traffic volume (Q), and the traveling speed (V). Although only the traffic counters 20a to 20c are shown in FIG. 1, other traffic counters may also be included. In the following description, the traffic counters 20a to 20c will be collectively referred to as "traffic counters 20."

[0018] The weather information table 30 includes weather information for various locations collected from information on the Japan Meteorological Agency website, etc. The estimation device 100, which will be described later, receives weather information from the weather information table 30.

[0019] The estimation device 100 has a comprehensive data table 141 and is a device that estimates the traffic volume in an observation area based on the comprehensive data table 141.

[0020] Fig. 2 is a diagram showing an example of the data structure of the coverage data table. As shown in Fig. 2, the coverage data table 141 has area identification information, time period, weather, traffic volume (Q), driving speed (V), and traffic density (K). The area identification information, time period, traffic volume (Q), and driving speed (V) are information included in the traffic counter information. The weather information is information included in the weather information table 30. The traffic density (K) is calculated based on equation (1).

[0021] Traffic volume (Q) = Traffic density (K) x Traveling speed (V) (1)

[0022] For example, the estimating device 100 executes a preparatory phase process and an estimation phase process. The preparatory phase process and the estimation phase process will be described below.

[0023] (Preparation phase processing) The following describes the processing of the preparation phase executed by the estimation device 100. For example, the estimation device 100 receives traffic counter information from the traffic counter 20, receives weather information from the weather information table 30, and calculates traffic density (K) based on the traffic volume (Q) and traveling speed (V) included in the traffic counter information and equation (1). The estimation device 100 registers in the exhaustive data table 141 a record that associates the area identification information, time period, traffic volume (Q), traveling speed (V) included in the traffic counter information, the calculated traffic density (K), and the weather received from the weather information table 30.

[0024] The estimation device 100 repeatedly executes the above process every time it receives traffic counter information from the traffic counter 20 .

[0025] The estimation device 100 generates a KV scatter diagram for each piece of area identification information based on the comprehensive data table 141. Fig. 3 is a diagram showing an example of a KV scatter diagram. For example, the KV scatter diagram G1 shown in Fig. 3 is the KV scatter diagram for the area identification information "observation area A1." The vertical axis of the KV scatter diagram G1 corresponds to the traveling speed (V), and the horizontal axis corresponds to the traffic density (K).

[0026] For example, the estimation device 100 identifies records with area identification information "observation area A1" from the comprehensive data table 141, and for each identified record, performs a process of plotting the relationship between driving speed (V) and traffic density (K), thereby generating a KV scatter plot G1.

[0027] The estimating device 100 generates KV scatter diagrams corresponding to other area identification information in the same manner as above.

[0028] After generating the KV scatter plot, the estimation device 100 performs two-dimensional kernel density estimation on the KV scatter plot. Kernel density estimation expresses the distribution of samples in the KV scatter plot as a continuous function. As a result of kernel density estimation, the kernel becomes high in areas where samples are concentrated and low in areas where samples are scattered.

[0029] FIG. 4 is a diagram showing an example of the results of two-dimensional kernel density estimation. FIG. 4 shows the results when two-dimensional kernel density estimation is performed on KV scatter plot G1. The results when two-dimensional kernel density estimation is performed on KV scatter plot G1 in FIG. 3 are denoted as "estimation result G1'." For example, in estimation result G1', areas with high kernel values ​​are displayed in dark colors, and areas with low kernel values ​​are displayed in light colors.

[0030] The estimation device 100 performs two-dimensional kernel density estimation on KV scatter diagrams corresponding to other area identification information in the same manner as described above.

[0031] After performing two-dimensional kernel density estimation, the estimation device 100 continuously changes the traffic density (K) for the estimation result and calculates a one-dimensional probability density function of the traveling speed (V) for each cross section of the traffic density (K).

[0032] FIG. 5 is a diagram for explaining the process of calculating a one-dimensional probability density function. In FIG. 5, the explanation will be given using the estimation result G1'. Also, in FIG. 5, as an example, the case where the traffic density (K) is changed to K = 40, 60, 80, and 120 will be explained. Graph G2 is a graph of the one-dimensional probability density function of the traveling speed (V) at various traffic densities (K). The vertical axis of graph G2 corresponds to the kernel density, and the horizontal axis corresponds to the traveling speed (V).

[0033] For example, the estimation device 100 calculates a one-dimensional probability density function of the traveling speed (V) for the cross section of the estimation result G1′ where K=40. Then, the relationship between the traveling speed (V) and the kernel density according to the probability density function is as shown by the curve 140.

[0034] The estimation device 100 calculates a one-dimensional probability density function of the traveling speed (V) for the cross section of the estimation result G1′ where K=60. Then, the relationship between the traveling speed (V) and the kernel density according to this probability density function is as shown by the curve 160.

[0035] The estimation device 100 calculates a one-dimensional probability density function of the traveling speed (V) for the cross section of the estimation result G1' at K=80. Then, the relationship between the traveling speed (V) and the kernel density according to this probability density function is as shown by the curve 180.

[0036] The estimation device 100 calculates a one-dimensional probability density function of the traveling speed (V) for the cross section of the estimation result G1′ at K=120. Then, the relationship between the traveling speed (V) and the kernel density according to this probability density function is shown by the curve 1120.

[0037] Similarly, for estimation results corresponding to other area identification information, the estimation device 100 continuously changes the traffic density (K) and calculates a one-dimensional probability density function of the traveling speed (V) for each cross section of the traffic density (K).

[0038] For example, the estimation device 100 stores information on the one-dimensional probability density function of each traffic density (K) corresponding to each area identification information in the probability density function table 142.

[0039] Fig. 6 is a diagram showing an example of the data structure of a probability density function table. As shown in Fig. 6, the probability density function table 142 associates area identification information with probability density function information. The area identification information is information for identifying an observation area. The probability density function information is information on a one-dimensional probability density function of each traffic density (K) corresponding to each area identification information.

[0040] (Estimation phase processing) Next, we will explain the processing of the estimation phase executed by the estimation device 100. When estimating the traffic density (K) of a certain observation area, the estimation device 100 uses vehicle travel information received from vehicles 10 traveling in the certain observation area and probability density function information corresponding to the area identification information of the certain observation area.

[0041] FIG. 7 is a diagram illustrating the processing of the estimation phase. For ease of explanation, the area identification information of a certain observation area is assumed to be "observation area A1." The estimation device 100 acquires probability density function information corresponding to the area identification information "observation area A1" from the probability density function table 142. The estimation device 100 plots a curve corresponding to the probability density function information (one-dimensional probability density function of traveling speed (V) at various traffic densities (K)) corresponding to "observation area A1" on graph G3 in FIG. 7. The vertical axis of graph G3 corresponds to the kernel density, and the horizontal axis corresponds to the traveling speed (V).

[0042] The curve l60 in Fig. 7 corresponds to the curve l60 described in Fig. 5. The curve l40 in Fig. 7 corresponds to the curve l40 described in Fig. 5. In Fig. 7, the curves l80 and l120 shown in Fig. 5 are omitted.

[0043] The estimation device 100 acquires multiple pieces of vehicle speed information from a vehicle 10 (multiple vehicles 10) traveling in an observation area corresponding to the area identification information "observation area A1," and plots data points on the axis of traveling speed (V) of graph G3. In the example shown in FIG. 7, plotted data points p1, p2, p3, p4, and p5 are shown. The distribution of data points p1, p2, p3, p4, and p5 is defined as a "speed information set."

[0044] The estimation device 100 determines the likelihood function when a speed information set is given as the product of the probability densities corresponding to each data point, and determines the traffic density (K) that gives the section with the largest value of the likelihood function as the traffic density (K) to be estimated. More specifically, the estimation device 100 executes the following process.

[0045] The estimating device 100 identifies an intersection p40-1 between a perpendicular line passing through the data point p1 and the curve l40. The estimating device 100 identifies an intersection p60-1 between a perpendicular line passing through the data point p1 and the curve l60.

[0046] The estimating device 100 identifies an intersection p40-2 between a perpendicular line passing through data point p2 and the curve l40. The estimating device 100 identifies an intersection p60-2 between a perpendicular line passing through data point p2 and the curve l60. The estimating device 100 identifies an intersection p40-3 between a perpendicular line passing through data point p3 and the curve l40. The estimating device 100 identifies an intersection p60-3 between a perpendicular line passing through data point p3 and the curve l60.

[0047] The estimating device 100 identifies an intersection p40-4 between a perpendicular line passing through data point p4 and the curve l40. The estimating device 100 identifies an intersection p60-4 between a perpendicular line passing through data point p4 and the curve l60. The estimating device 100 identifies an intersection p40-5 between a perpendicular line passing through data point p5 and the curve l40. The estimating device 100 identifies an intersection p60-5 between a perpendicular line passing through data point p3 and the curve l60.

[0048] The estimation device 100 calculates the value of the likelihood function for the curve l40 by multiplying the value of the intersection p40-1 (kernel density, the same applies below), the value of the intersection p40-2, the value of the intersection p40-3, the value of the intersection p40-4, and the value of the intersection p40-5.

[0049] The estimation device 100 calculates the value of the likelihood function for the curve l60 by multiplying the value of the intersection p60-1 (kernel density, the same applies below), the value of the intersection p60-2, the value of the intersection p60-3, the value of the intersection p60-4, and the value of the intersection p60-5.

[0050] Although not shown in Figure 7, the estimation device 100 similarly calculates likelihood function values ​​for curves obtained by calculating one-dimensional probability density functions of travel speed (V) for all cross sections obtained by varying the traffic density (K), which is a continuous variable in Figure 4, including curves l80 and l120. The estimation device 100 determines the traffic density (K) corresponding to the curve with the maximum likelihood function among the likelihood function values ​​for curves corresponding to all continuous traffic densities as the traffic density (K) to be estimated. For example, when the likelihood function value for curve l60 is maximum, the estimation device 100 determines the traffic density to be estimated as K = 60.

[0051] (Configuration of Estimation Device 100) Next, a description will be given of an example of the configuration of the estimation device 100 shown in Fig. 1, which executes the above-mentioned preparatory phase processing and estimation phase processing. Fig. 8 is a functional block diagram showing the configuration of the estimation device.

[0052] For example, the estimation device 100 includes a communication unit 110, an input unit 120, a display unit 130, a storage unit 140, and a control unit 150.

[0053] The communication unit 110 transmits and receives information to and from the vehicle 10, the traffic counter 20, and the weather information table 30 via the network 5. For example, the communication unit 110 is realized by a NIC (Network Interface Card) or the like.

[0054] The input unit 120 inputs various information to the control unit 150 of the estimation device 100. For example, the input unit 120 is realized by a keyboard, a mouse, a touch panel, etc. The user may operate the input unit 120 to specify an area for which traffic density (K) is to be estimated.

[0055] The display unit 130 displays information output from the control unit 150. For example, the display unit 130 is realized by a display or the like.

[0056] The storage unit 140 is a storage device that stores various types of information, and is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. For example, the storage unit 140 includes an exhaustive data table 141, a probability density function table 142, etc.

[0057] The coverage data table 141 is the coverage data table 141 described in Fig. 2. The coverage data table 141 has area identification information, time period, weather, traffic volume (Q), driving speed (V), traffic density (K), and the like.

[0058] The probability density function table 142 is the probability density function table 142 described in Fig. 6. The probability density function table 142 associates area identification information with probability density function information.

[0059] The control unit 150 is realized by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) executing various programs stored in a storage device inside the estimation device 100 using a RAM or the like as a working area. The control unit 150 may also be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0060] The control unit 150 has a registration unit 151, a probability density estimation unit 152, a calculation unit 153, and a traffic density estimation unit 154. For example, the processing of the registration unit 151, the processing of the probability density estimation unit 152, and the processing of the calculation unit 153 correspond to the processing of the preparation phase described above. The processing of the traffic density estimation unit 154 corresponds to the processing of the estimation phase described above.

[0061] When the registration unit 151 receives traffic counter information from the traffic counter 20, it calculates the traffic density (K) based on the traffic volume (Q) and traveling speed (V) included in the traffic counter information and equation (1). The registration unit 151 registers in the coverage data table 141 a record that associates the area identification information, time period, traffic volume (Q), traveling speed (V) included in the traffic counter information with the traffic density (K) that is the calculation result and the weather included in the weather information table 30.

[0062] The registration unit 151 repeatedly executes the above process every time it receives traffic counter information from the traffic counter 20.

[0063] The probability density estimation unit 152 generates a KV scatter diagram for each piece of area identification information based on the comprehensive data table 141. An example of the KV scatter diagram generated by the probability density estimation unit 152 is the KV scatter diagram G1 shown in FIG.

[0064] After generating the KV scatter diagram, the probability density estimation unit 152 estimates a two-dimensional continuous probability density function for the KV scatter diagram. For example, when estimating a two-dimensional continuous probability density function, the probability density estimation unit 152 performs kernel density estimation as described in FIG. 4.

[0065] The probability density estimation unit 152 outputs to the calculation unit 153 the estimation results of the two-dimensional continuous probability density function executed for each KV scatter diagram for each piece of area identification information.

[0066] The calculation unit 153 continuously changes the traffic density (K) for the estimation result of the two-dimensional continuous probability density function, and calculates the one-dimensional probability density function of the traveling speed (V) for the cross section of each traffic density (K). The process by which the calculation unit 153 calculates the one-dimensional probability density function of the traveling speed (V) is the same as that described in FIG. 5.

[0067] The calculation unit 153 stores information on the one-dimensional probability density function of each traffic density (K) corresponding to each area identification information in the probability density function table 142.

[0068] The traffic density estimation unit 154 receives, from the input unit 120, etc., the designation of area identification information of the observation area for which traffic density (K) is to be estimated. For convenience of explanation, the following description will be given assuming that "observation area A1" is designated as the area identification information.

[0069] The traffic density estimation unit 154 acquires probability density function information corresponding to the area identification information "observation area A1" from the probability density function table 142. As described with reference to graph G3 in Fig. 7 , the traffic density estimation unit 154 calculates the probability density function information (one-dimensional probability density function of traveling speed (V) at various traffic densities (K)) corresponding to the "observation area A1."

[0070] The traffic density estimation unit 154 acquires multiple vehicle travel speed information from a vehicle 10 (multiple vehicles 10) traveling in the observation area corresponding to the area identification information "observation area A1" and plots data points on the travel speed (V) axis of graph G3.

[0071] The traffic density estimation unit 154 calculates the likelihood function when a speed information set is given as the product of the probability densities corresponding to each data point, and estimates the traffic density (K) that gives the cross section with the largest value of the likelihood function as the traffic density (K) of the estimation target (the observation area corresponding to the area identification information "observation area A1"). The traffic density estimation unit 154 outputs the estimation result to the display unit 130 for display.

[0072] The rest of the explanation of how the estimation device 100 estimates the traffic density (K) is the same as the explanation given in FIG.

[0073] (Processing procedure of the estimation device 100) Next, an example of the processing procedure of the estimation device 100 will be described. Fig. 9 is a flowchart showing the processing procedure of the preparation phase. As shown in Fig. 9, the registration unit 151 of the estimation device 100 receives traffic counter information from the traffic counter 20 and weather information from the weather information table 30 (step S101).

[0074] The registration unit 151 calculates the traffic density (K) based on the traffic volume (Q) and traveling speed (V) included in the traffic counter information and equation (1) (step S102). The registration unit 151 registers the record in the coverage data table 141 (step S103).

[0075] The probability density estimation unit 152 of the estimation device 100 acquires the traffic density (K) and the traveling speed (V) from the exhaustive data table 141 and generates a KV scatter diagram (step S104). The probability density estimation unit 152 estimates a two-dimensional continuous probability density function for the KV scatter diagram (step S105).

[0076] The calculation unit 153 of the estimation device 100 continuously changes the traffic density (K) for the estimation result of the two-dimensional continuous probability density function, and calculates the one-dimensional probability density function of the traveling speed (V) for the cross section of each traffic density (K) (step S106).

[0077] The calculation unit 153 stores information on the one-dimensional probability density function of each traffic density (K) in the probability density function table 142 (step S107).

[0078] 10 is a flowchart showing the processing procedure of the estimation phase. As shown in Fig. 10, the traffic density estimation unit 154 of the estimation device 100 receives designation of area identification information of an observation area for which traffic density (K) is to be estimated (step S201).

[0079] The traffic density estimation unit 154 acquires probability density function information corresponding to the specified area identification information from the probability density function table 142 (step S202). Based on the probability density function information, the traffic density estimation unit 154 calculates a one-dimensional probability density function of the traveling speed (V) at various traffic densities (K) (step S203).

[0080] The traffic density estimation unit 154 acquires a plurality of pieces of vehicle speed information from vehicles 10 traveling in the observation area corresponding to the area identification information (step S204). The traffic density estimation unit 154 estimates the traffic density (K) of the observation area corresponding to the area identification information based on a one-dimensional probability density function of the traveling speed (V) at various traffic densities (K) and the plurality of pieces of vehicle speed information (step S205). The traffic density estimation unit 154 outputs the estimation result (step S206).

[0081] (Effects of the embodiment) According to the embodiment, in the preparation phase, the estimation device 100 registers data on road conditions in the real environment in the comprehensive data table 141, and generates information on a one-dimensional probability density function of each traffic density (K) corresponding to each area identification information based on the comprehensive data table 141. In the estimation phase, the estimation device 100 receives the distribution of traveling speeds from multiple vehicles traveling in the target observation area, and estimates the traffic density of the target observation area based on the multiple one-dimensional probability density functions and traveling speed distributions generated in the preparation phase. This enables accurate estimation of the traffic volume of roads in the real environment.

[0082] Furthermore, the estimation device 100 treats the data in the exhaustive data table 141 as a continuous distribution, and determines the traffic density (K) that gives the cross section with the largest value of the likelihood function as the traffic density (K) to be estimated. This ensures the plausibility of the estimation result even when the number of travel speed data obtained from vehicles is small.

[0083] (Configuration of Estimation Device 100) Each component of the estimating device 100 is a functional concept and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of the functions of the estimating device 100 is not limited to that shown in the figure, and all or part of the functions can be functionally or physically distributed or integrated in any unit depending on various loads, usage conditions, etc.

[0084] Furthermore, all or any part of the processes performed in the estimation device 100 may be realized by a CPU and a program analyzed and executed by the CPU. Furthermore, each process performed in the estimation device 100 may be realized as hardware using wired logic.

[0085] Furthermore, among the processes described in the embodiments, all or part of the processes described as being performed automatically can be performed manually. Alternatively, all or part of the processes described as being performed manually can be performed automatically using a known method. In addition, the processing procedures, control procedures, specific names, and information including various data and parameters described above and illustrated can be changed as appropriate unless otherwise specified.

[0086] (Other processes performed by the estimation device 100) The above-described estimation device 100 generates a KV scatter diagram for each area identification information based on the comprehensive data table 141, performs two-dimensional kernel density estimation on the KV scatter diagram for each area identification information, continuously changes the traffic density (K) for the kernel density estimation for each area identification information, and calculates a one-dimensional probability density function of the traveling speed (V) for a cross section of each traffic density (K). In other words, the estimation device 100 calculates the one-dimensional probability density function of the traveling speed (V) based on information "for each area identification information," but the present invention is not limited to this.

[0087] By performing the above processing for "each set of area identification information and weather," "each set of area identification information and time zone," or "each set of area identification information, weather, and time zone," the estimation device 100 can calculate a one-dimensional probability density function of the traveling speed (V) for "each set of area identification information and weather," "each set of area identification information and time zone," or "each set of area identification information, weather, and time zone." This makes it possible to estimate traffic volume that matches actual traffic conditions formed by a complex combination of the effects of weather, time, etc.

[0088] For example, when the estimation device 100 generates a KV scatter diagram for each combination of area identification information "observation area A1" and weather "fine," the estimation device 100 extracts records corresponding to each combination of area identification information "observation area A1" and weather "fine" from the comprehensive data table 141, and generates the KV scatter diagram. The process from when the estimation device 100 generates the KV scatter diagram to when it calculates the one-dimensional probability density function of the traveling speed (V) is the same as the process described above.

[0089] When the estimation device 100 generates a KV scatter diagram for each combination of area identification information "observation area A1" and time period "T1", the estimation device 100 extracts records corresponding to each combination of area identification information "observation area A1" and time period "T1" from the comprehensive data table 141, and generates the KV scatter diagram. The process performed by the estimation device 100 from generating the KV scatter diagram to calculating the one-dimensional probability density function of the traveling speed (V) is the same as the process described above.

[0090] When the estimation device 100 generates a KV scatter diagram for each combination of area identification information "observation area A1", weather "fine", and time period "T1", the estimation device 100 extracts records corresponding to each combination of area identification information "observation area A1", weather "fine", and time period "T1" from the comprehensive data table 141, and generates a KV scatter diagram. The process performed by the estimation device 100 from generating the KV scatter diagram to calculating the one-dimensional probability density function of the traveling speed (V) is the same as the process described above.

[0091] Furthermore, the estimation device 100 displays the estimation results on the display unit 130, but this is not limiting. For example, the traffic density estimation unit 154 of the estimation device 100 may transmit information associating the estimation results with area identification information to a server that manages the traffic volume (congestion status) of each observation area.

[0092] [program] 11 is a diagram showing an example of a computer in which the estimation device 100 is realized by executing a program. The computer 1000 includes, for example, a memory 1010 and a CPU 1020. The computer 1000 also includes a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.

[0093] The memory 1010 includes a ROM 1011 and a RAM 1012. The ROM 1011 stores a boot program such as a BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to a hard disk drive 1031. The disk drive interface 1040 is connected to a disk drive 1041. A removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1041. The serial port interface 1050 is connected to a mouse 1051 and a keyboard 1052, for example. The video adapter 1060 is connected to a display 1061, for example.

[0094] The hard disk drive 1031 stores, for example, an OS 1091, an application program 1092, a program module 1093, and program data 1094. That is, a program that defines each process of the estimating device 100 is implemented as a program module 1093 in which code executable by the computer 1000 is written. The program module 1093 is stored, for example, in the hard disk drive 1031. For example, the program module 1093 for executing processes similar to those of the functional configuration of the estimating device 100 is stored in the hard disk drive 1031. Note that the hard disk drive 1031 may be replaced by an SSD (Solid State Drive).

[0095] Furthermore, setting data used in the processing of the above-described embodiment is stored as program data 1094, for example, in memory 1010 or hard disk drive 1031. Then, CPU 1020 reads out program module 1093 and program data 1094 stored in memory 1010 or hard disk drive 1031 into RAM 1012 as necessary and executes them.

[0096] The program module 1093 and program data 1094 are not limited to being stored in the hard disk drive 1031, but may also be stored in, for example, a removable storage medium and read by the CPU 1020 via the disk drive 1041 or the like. Alternatively, the program module 1093 and program data 1094 may be stored in another computer connected via a network (such as a LAN (Local Area Network) or WAN (Wide Area Network)). The program module 1093 and program data 1094 may then be read by the CPU 1020 from the other computer via the network interface 1070.

[0097] Although the present invention has been described above as an embodiment, the present invention is not limited to the description and drawings that form part of the disclosure of the present invention. In other words, other embodiments, examples, and operational techniques that can be made by those skilled in the art based on the present invention are all included in the scope of the present invention. [Explanation of symbols]

[0098] 100 Estimator 110 Communications Department 120 Input section 130 Display section 140 Storage section 141 Comprehensive Data Tables 142 Probability Density Function Table 150 control section 151 Registration Department 152 Probability density estimation unit 153 Calculation Department 154 Traffic density estimation section

Claims

1. a probability density estimation unit that estimates a two-dimensional continuous probability density function from a distribution of samples that indicate the relationship between traffic density and traveling speed for a plurality of vehicles that have traveled in a target area in the past; a calculation unit that varies traffic density, which is a continuous variable, with respect to the two-dimensional continuous probability density function and calculates one-dimensional probability density functions of traveling speeds corresponding to each traffic density; a traffic density estimation unit that receives a distribution of traveling speeds from a plurality of vehicles traveling in the target area and estimates a traffic density in the target area based on a one-dimensional probability density function and the distribution of the traveling speeds; An estimation device comprising:

2. The estimation device according to claim 1, characterized in that the traffic density estimation unit calculates a product of probability densities corresponding to a plurality of traveling speeds included in the traveling speed distribution based on the one-dimensional probability density function, and estimates the traffic density based on the product of the one-dimensional probability density function.

3. 2. The estimation device according to claim 1, wherein the probability density estimation unit estimates a two-dimensional kernel function from a distribution of samples indicating a relationship between traffic density and traveling speed for a plurality of vehicles.

4. An estimation method executed by an estimation device, A step of estimating a two-dimensional continuous probability density function from a distribution of samples showing the relationship between traffic density and traveling speed for a plurality of vehicles that have previously traveled in a target area; a step of varying traffic density, which is a continuous variable, for the two-dimensional continuous probability density function and calculating one-dimensional probability density functions of travel speeds corresponding to each traffic density; receiving a distribution of travel speeds from a plurality of vehicles traveling in the target area, and estimating a traffic density in the target area based on a one-dimensional probability density function and the distribution of travel speeds; An estimation method comprising:

Images